Geosemantic Snapping for Sketch‐Based Modeling

Abstract

Figure 1: Using our method, the human modeler simply drags-and-drops the shape primitives into approximate position; the system provides real-time precise geometric snapping feedback and infers geosemantic constraints between the primitives. (Total modeling time: one minute.) Abstract Modeling 3D objects from sketches is a process that requires several challenging problems including segmentation, recognition and reconstruction. Some of these tasks are harder for humans and some are harder for the machine. At the core of the problem lies the need for semantic understanding of the shape's geometry from the sketch. In this paper we propose a method to model 3D objects from sketches by utilizing humans specifically for semantic tasks that are very simple for humans and extremely difficult for the machine, while utilizing the machine for tasks that are harder for humans. The user assists recognition and segmentation by choosing and placing specific geometric primitives on the relevant parts of the sketch. The machine first snaps the primitive to the sketch by fitting its projection to the sketch lines, and then improves the model globally by inferring geosemantic constraints that link the different parts. The fitting occurs in real-time, allowing the user to be only as precise as needed to have a good starting configuration for this non-convex optimization problem. We evaluate the accessibility of our approach with a user study.

Meta-representation of shape families

“…However, in the shape editing approaches discussed above, structure is typically learned from a single shape, and not inferred from a set of shapes as is our goal in this work. Detecting structural relationships between multiple components is also important in other domains : Fisher et al [2011] detect relationships between objects in scenes to perform scene or object retrieval; Yumer and Kara [2012] produce identity preserving mutually consistent coabstractions for shape collections; while Shtof et al [2013] apply the structure-aware principle to sketch-based modeling. Differently from these works, Sumner et al [2005] deform a mesh with inverse kinematics according to example deformations.…”

Basic Visual Disciplines in Heritage Conservation: Outline of Selected Perspectives in Teaching and Learning

“…Therefore, the main aims of sketch based modeling are seen as to automate or assist the sketch-to-3D translation process (Olsen et al 2009), to bridge the gap between concept design and computer-based modeling programs (Kondo 2009). According to Shtof et al (2013), it would be nice if simple 3D models could be created easily from 2D sketches for quick previews and as a starting point for further editing and creation of more complex models.…”

“…Professional 3D modelers nearly always begin the modeling process by sketching, either on paper or in a 2D sketching application (Shtof et al 2013). Frequently, people try to explain their ideas through a 2D sketch.…”

Design and fabrication by example

“…First, we select the leaf nodes and assign their degrees of freedom q i and deformation functions F i . Second, we analyze the semantic geometric (geosemantic [Shtof et al 2013]) relationships between parts of our model in order to define structure-preserving constraints at each level of the hierarchy, thus determining A i (see example in Figure 5). …”